A single demyelinating attack is enough to limit brain growth in children.

Acquired demyelinating syndromes (ADS) represent acute neurologic illnesses characterized by deficits persisting for at least 24 hours and involving the optic nerve, brain, or spinal cord, associated with regional areas of increased T2 signal on conventional MRI. In contrast to the 75% of adult patients presenting with clinically isolated syndrome (CIS) who are eventually diagnosed with clinically defined multiple sclerosis (MS), the majority of children presenting with ADS have a monophasic illness with a good overall prognosis: more than 90% of children achieve a full neurologic recovery. Over the last decade, neuroimaging studies have focused on differentiating between MS and its mimics, and correlating radiologic measures with clinical outcome and neuropsychological testing in pediatric MS. Quantitative MRI analysis in children with MS has demonstrated that children fail to achieve the expected brain growth trajectory and show a reduced whole brain volume in adolescence; the regions more affected are the gray matter and the thalamus. The reduction of brain growth due to MS has been attributed to CNS inflammation, demyelination, and neurodegeneration. Brain atrophy, as measured by MRI, particularly in adults, may reflect gray matter neurodegeneration more than CNS active inflammation. No overlap between double inversion recovery detectable gray matter lesions and gray matter atrophy has been detected on MRI. Assessing brain atrophy in children is challenging; particular attention is required to take the typical expected developmental trajectory and correct for head size and age. In adultonset MS, the total intracranial volume (gray matter, white matter, and CSF) is considered insensitive to atrophy; thus, the ratio of brain parenchymal volume (white and gray matter) to the total brain volume within the brain surface contour is used to measure whole brain atrophy. However, skull growth further complicates the interpretation of volume changes over time in children, which necessitates normalization of brain size to skull size in the pediatric population. In this issue of Neurology®, Aubert-Broche et al. report the prospective analysis of serial brain imaging in 83 children with monophasic ADS identified from the Canadian Pediatric Demyelinating Disease Study. These included children with acute disseminated encephalomyelitis (ADEM), CIS with brain involvement, and optic neuritis or transverse myelitis with a normal intracranial MRI. Although a relapsing disease course cannot be completely ruled out, the median period of observation of 4 years makes this unlikely. All children were invited to attend scans clinically at onset, and a subset research scans at approximately 2 weeks, then 3, 6, and 12 months, and annually thereafter. However, the numbers of children scanned at the prespecified prospective time points are not provided. Normative data were used to calculate ageand sex-specific z scores. This was particularly important in the pediatric population to control for intersubject developmental variability, and enabled the comparison of patients with different ages at onset, different stages of brain maturation at first measurement, and effect of disease duration on brain structures independent of age. The findings presented by Aubert-Broche et al. demonstrate that children with monophasic demyelination have reduced age-expected brain growth, driven via reduced white matter growth. Patients with ADEM showed the most deviation from age-expected trajectory. Interestingly, unlike adults with CIS, in which atrophy rates are not different from healthy controls, reduction in age-expected brain growth was seen even in children with ADS without brain involvement on conventional MRI. In contrast with previous observations in adults and children with MS, the thalamus was not preferentially affected compared to the whole brain. Aquaporin-4 antibodies were negative in the 63 children tested, but myelin oligodendrocyte glycoprotein antibodies (MOG-Ab) were not tested in this cohort. MOG-Ab occur in up to 50% of children with ADS and are associated with a non-MS disease course. It cannot be excluded that a proportion